wechat-article-extractor

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill relies on npx wechat-article-extractor to perform its primary function. This command downloads and executes a package from the npm registry at runtime. While this is the standard distribution method for the author's tool, it involves executing remote code.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it parses content from external, untrusted WeChat articles (mp.weixin.qq.com). Malicious actors could embed instructions in these articles to influence the agent's behavior.
  • Ingestion points: External web content retrieved from mp.weixin.qq.com and weixin.sogou.com via the extract function and CLI tool.
  • Boundary markers: The skill documentation does not suggest using delimiters or specific system instructions to isolate the extracted article content from the agent's control logic.
  • Capability inventory: The skill performs network operations to fetch data and outputs processed text/JSON to the agent, which may then be used for further decision-making.
  • Sanitization: There is no evidence of prompt-specific sanitization; the mentioned lodash.unescape is used for processing HTML entities rather than security filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 05:48 AM